JP6716334B2 - Information processing apparatus, information processing method, and computer program. - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and a computer program that process a signal obtained by measuring a moving body.

In general, an optical rotary encoder is widely used to detect the movement amount, angle, position, etc. of a moving body such as a motor. In an optical rotary encoder, a light emitting element such as a light emitting diode or an LED irradiates a rotating signal plate on which a pattern required for signal detection is formed with uniform light. Then, the light transmitted through the pattern or the light reflected from the signal plate is detected by a light receiving element such as a photodiode or a phototransistor. An electric signal pattern is created from the detection result. Then, the encoder output signal is generated based on the electric signal pattern. The encoder output signal is known to be functionally classified into an incremental system that outputs a relative position and an absolute system that outputs an absolute position. Here, the incremental method will be described.

In the incremental method, a technique for realizing highly accurate position detection by interpolation processing is known. This is based on the precondition that the amplitude and offset values of the encoder signal are uniform and two-phase analog sine wave signals whose phases are different by 90° are output.

As a specific method of interpolation processing, a method using resistance division and a method using arctangent calculation are known.

As a method of performing an arctangent calculation, a rotation calculation processing method of coordinate data known as a “CORDIC (COordinate Rotation Digital Computer) algorithm” is known. In CORDIC, first, an input signal is treated as two-dimensional coordinate data. Then, the arctangent calculation can be realized by executing the simple rotation of bit shift and addition/subtraction for the coordinate rotation of the two-dimensional coordinate data.

However, the actual encoder output signal contains harmonic components and is not an ideal sine wave signal. Therefore, even if the error of the amplitude, the offset, and the phase of the encoder output signal is corrected, the sine wave signal is not strict, and a detection error occurs when performing the interpolation process.

In the technique disclosed in Patent Document 1, in order to correct such a detection error, a value obtained by differentiating the detected displacement amount (detected angle) of the encoder, that is, a correction value that makes the displacement speed constant is generated. ing. The detected position is corrected by setting the generated correction value for each detected position. Further, in the technique disclosed in Patent Document 2, the transition of the detection error is periodic, and the detection error amount is sequentially calculated by approximating the detection error amount with a sine wave to detect the detection position. Is being corrected.

JP, 2009-303358, A JP 2006-170837 A

However, for example, in the method disclosed in Patent Document 1, it is necessary to prepare the correction value for the resolution of the detection position. Therefore, the memory area for storing the correction value becomes enormous as the resolution of the detection position becomes higher. On the other hand, in Patent Document 2, first, a calculation time is required to generate the correction value. Further, even if the transition of the detection error can be approximated to an ideal sine wave, the position information (reference position) used to calculate the detection error is the detection position including the error, so that the theoretical Moreover, the calculation accuracy of the detection error becomes low.

The present invention has been made in view of the above problems, and an object thereof is to process an encoder signal of a moving body with high accuracy while suppressing a memory capacity for correcting a detection error.

In order to solve the above-mentioned problems, the information processing apparatus of the present invention is an acquisition unit that acquires a first-phase analog wavy signal and a second-phase analog wavy signal obtained by measuring a moving body. And a second cosine wave that is a higher harmonic than the first cosine wave and the first cosine wave with respect to the first phase analog wavy signal, to the second phase analog wavy signal On the other hand, the information of the phase satisfying the relational expression of the arctangent using the first sinusoidal wave and the second sinusoidal wave which is a higher harmonic wave than the first sinusoidal wave is converted into the arctangent relational expression by iterative calculation. And a calculating unit that calculates by converging the included rotation angle.

According to the present invention, it is possible to process an encoder signal of a moving object with high accuracy while suppressing a memory capacity for correcting a detection error.

3 is a control block diagram of the motor control device according to the first embodiment. FIG. FIG. 3 is a diagram showing a configuration of a displacement detection unit according to the first embodiment. The figure which shows the structure of the interpolation process part which concerns on 1st Embodiment. The figure which shows the structure of the arctangent calculation part which concerns on 1st Embodiment. The figure which shows the structure of the arctangent calculation part which concerns on 2nd Embodiment. The figure which shows the structure of the arctangent calculation part which concerns on 3rd Embodiment. The figure which shows the structure of the arctangent calculation part which concerns on 4th Embodiment. The figure which shows the structure of the displacement detection part which concerns on 5th Embodiment. The figure which shows the structure of the harmonic component removal filter part which concerns on 5th Embodiment. The schematic diagram of a motor control device. The figure explaining a 2-phase encoder output signal. The figure which shows an ideal encoder signal waveform (sine wave) and an actual encoder signal waveform as an example. The figure which shows the detection displacement amount and detection error with respect to the actual displacement of a motor as an example. The figure which shows the harmonic distortion encoder signal waveform in 1st Embodiment, 2nd Embodiment, and 5th Embodiment. The figure which shows the calculation result in 1st Embodiment. The figure which shows the calculation result in 2nd Embodiment. The figure which shows the calculation result when there is no correction in 3rd Embodiment. The figure which shows the harmonic distortion encoder signal waveform in 3rd Embodiment. The figure which shows the calculation result in 3rd Embodiment. The figure which shows the calculation result in 5th Embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

(First embodiment)
As a first embodiment of an information processing apparatus according to the present invention, an information processing apparatus mounted on a motor control device and capable of removing harmonic distortion superimposed on an encoder signal will be described.

(Structure of motor controller)
FIG. 10 is a schematic diagram of the motor control device according to the present embodiment. 10A shows an overall view of the motor control device 1000, and FIG. 10B shows a schematic plan view of the encoder scale 1002. Further, the motor control device 1000 detects the rotation angle (displacement amount) of the motor shaft of the rotary motor 1001.

The detection is performed by an encoder scale 1002 having a rotary slit disc and a fixed slit disc, and a sensor unit 1003 having a light emitting element (light emitting diode) and a light receiving element (photodiode). The rotary slit disk is attached to the motor shaft of the rotary motor 1001 and rotates with the motor shaft. On the other hand, the fixed slit disk and the sensor unit 1003 are fixed to the substrate. A rotary slit disk and a fixed slit disk are arranged between the light emitting element and the light receiving element.

The encoder scale 1002 is provided with a plurality of slits 1005. The encoder scale 1002 rotates around the motor shaft as a rotation axis center in accordance with the rotational displacement of the motor shaft of the rotary motor 1001. The light of the light emitting element is transmitted or blocked by the rotation of the rotary slit disk. Further, the fixed slit disk is divided into two fixed slits in order to make the output signals of the encoder into two phases.

The sensor unit 1003 is provided with two light receiving elements, and each of the light receiving elements passes through the slit 1005 and further detects the light divided into two kinds of the A phase pattern and the B phase pattern by the fixed slit. Thereby forming a two-phase electric signal pattern.

FIG. 11 is a diagram illustrating a two-phase encoder output signal. As shown in FIG. 11, the optical encoder generates an A-phase signal and a B-phase signal whose phases are different from each other by 90°. These are cases where a two-phase encoder signal is obtained, but when obtaining a two-phase encoder signal, a plurality of fixed slits and light receiving elements are prepared. The detection principle of the encoder is not limited to the optical type (transmissive type), and other types such as an optical type using reflected light, an electrostatic type, and a magnetic type can be adopted. Further, the rotary slit disk may be attached to a driven body driven by the motor control device 1000, instead of the motor shaft, to detect the amount of movement of the driven body.

The motor controller 1004 controls the rotary drive of the rotary motor 1001. The motor controller 1004 includes a drive unit that drives the rotary motor 1001 and a control unit that controls the drive unit. The motor controller 1004 compares the motor rotation angle (target displacement amount), which is a target value, with the motor detection angle (detection displacement amount), which is an actual measurement value. Then, feedback control is performed so that the measured value becomes equal to the target value. This is a motor control method called closed loop, which is used even when the motor detection angle (detection displacement amount) is acquired, but the open loop control method drives the motor without comparing it with the target value. It is possible.

FIG. 1 is a control block diagram of the motor control device according to the present embodiment. The drive unit 101 supplies a predetermined drive signal 1 to the rotary motor 1001 based on an output signal from the motor controller 1004 which is a higher-order control unit. The motor shaft, which is the movable portion 102 (object to be measured), is displaced by a predetermined rotation angle when the drive signal 1 is input. The displacement detection unit 103 detects the displacement amount (positional displacement amount) of the rotation angle of the movable unit 102 and outputs the detection result.

FIG. 2 is a diagram showing the configuration of the displacement detection unit 103 in this embodiment. The encoder 201 interlocked with the movable unit 102 generates an encoder signal of a sinusoidal two-phase signal having a 90-degree phase difference as shown in FIG. 11 according to the displacement amount, and outputs the encoder signal to the interpolation processing unit 202. As shown in the actual encoder signal of FIG. 12 (only one phase is shown), this encoder signal is not an ideal sine wave, but generally contains a harmonic component. This harmonic component is generated by a complicated combination of optical phenomena such as generation of light intensity distribution of the light emitting element and diffraction by the fixed slit and the rotating slit. Here, as the encoder 201, for example, an optical encoder is used. However, it is known that the phenomenon in which harmonic components are generated and superposed occurs also in a magnetic encoder due to magnetic field distribution distortion, magnetoresistive distortion, etc., and is not limited to an optical encoder. ..

FIG. 3 is a diagram showing a configuration of the interpolation processing unit 202. The analog/digital (A/D) converter 301 converts the analog A-phase signal 4 and the analog B-phase signal 5 that are analog wave-shaped signals into digital signals (digital A-phase signal 6 and digital B-phase signal 7). belongs to. A filter circuit and a noise removing circuit are mounted on the interpolation processing unit 202 in order to remove a noise component included in the encoder signal before and after the A/D conversion. The filter circuit and the noise removing circuit may have known configurations, and thus detailed description thereof will be omitted.

The offset/phase difference correction unit 302 performs offset processing on each of the reference values of the digital A-phase signal 6 and the digital B-phase signal 7. By this offset processing, the A-phase and B-phase encoder signals become positive and negative signals with 0 as a reference.

Further, in the phase difference correction, the correction processing is performed so that the phase difference between the A phase and the B phase becomes 90°. In this phase difference correction processing, the rotation calculation processing is performed on one encoder signal by the amount of error from the phase difference of 90°, and the sum of A phase and B phase and the difference between A phase and B phase are newly added. By using a phase signal, there is a method of guaranteeing a 90° phase for a new two-phase signal. In this way, the adjusted digital A-phase signal 8 and the adjusted digital B-phase signal 9 which are offset/phase difference corrected are obtained.

The area determination unit 303 determines the quadrant (area) of the Lissajous made from the adjusted digital A-phase signal 8 and the adjusted digital B-phase signal 9. The first quadrant when both the A-phase signal and the B-phase signal are positive, the second quadrant when the A-phase signal is negative, the second quadrant when the B-phase signal is positive, and the third quadrant when both the A-phase signal and the B-phase signal are negative In the quadrant, when the A-phase signal is positive and the B-phase signal is negative, the fourth quadrant is used. At this time, the position information of the encoder is incremented or decremented as the quadrant moves, and rough position information is obtained. For example, when the encoder slit interval is defined as 360°, this position information can acquire the position in 90° units. The sum of this rough position information and the interpolation position to be described later is the detected displacement amount. However, in this description, since only the interpolation process (interpolation position) is focused on, this rough position information will be described below. Is omitted.

Further, in order that the arithmetic processing after the area determination unit 303 can be executed only with a positive value, the area determination unit 303 uses the A phase data X10 that is the absolute value of the adjusted digital A phase signal and the adjusted digital B phase signal. B-phase data Y11 that is the absolute value of is output.

The arctangent calculation unit 304 performs arctangent value (arctan) calculation based on the A-phase data X10 and the B-phase data Y11 to calculate the interpolated position, and 0° to 90°, that is, within one quadrant. The interpolation position 12 is output.

The angle conversion unit 305 corrects the interpolation position 12 in the range of 0° to 90° calculated by the arctangent calculation unit 304 by the quadrant information (four divisions) by the region determination unit 303. Specifically, in the first quadrant and the third quadrant, the interpolated position 12 is the detected displacement amount as it is, and in the second quadrant and the fourth quadrant, 90°-the interpolated position is the detected displacement amount. .. (When considering the rough position information, the rough position information is added here) In this way, the displacement detection unit 103 calculates a highly accurate angle value that has been subjected to interpolation processing from the encoder signal. Here, if the calculation of the angle by the arctangent calculation unit 304 is performed on the assumption that the sine wave is an ideal as in the conventional technique, a calculation error occurs. This is due to the harmonic components included in the encoder signal, which causes a detection error.

FIG. 12 is a diagram showing an ideal encoder signal waveform (sine wave) and an actual encoder signal waveform. Further, FIG. 13 is a diagram showing the detected displacement amount and the detection error when the actual encoder signal waveform shown in FIG. 14 is input. The detection error is the difference between the actual displacement amount of the driven motor and the detected displacement amount.

As shown in FIG. 13, the detected displacement amount is not linear with the actual displacement amount due to the influence of the detection error. That is, it is a curve on the graph. When the encoder signal is an ideal sine wave, no displacement detection error occurs and the actual displacement amount and the detected displacement amount are linear. In other words, it becomes a straight line on the graph.

Further, as can be seen from the change in the detection error shown in FIG. 13, the detection error periodically depends on the actual displacement amount, that is, depends on the generation amount of the harmonic component of the encoder signal, and Within one encoder slit interval (360 [deg]), fluctuations of four cycles occur. Then, an error of about ±5 [deg] occurs. However, this is an example of characteristics when the encoder signal has the encoder waveform shown in FIG. 14, and is not limited to this. The above tendency also applies to the cosine wave.

As described above, a harmonic component is generated in the encoder signal, and a detection error will occur if an operation that does not consider the harmonic component is performed. In the present embodiment, in order to improve the detection accuracy of the displacement detection unit 103, the arctangent calculation (harmonic distortion correction CORDIC rotation calculation method) is performed in consideration of the harmonic components included in the encoder signal. The details will be described below.

(Structure of arctangent calculation unit considering harmonic components)
FIG. 4 shows a configuration diagram of the arctangent calculation unit 304 using the CORDIC rotation calculation capable of correcting harmonic distortion in the present embodiment. It should be noted that although the present embodiment has been described using the CORDIC operation, the rotation angle may be converged and obtained using an iterative calculation such as a binary search. The A-phase input signal X (A-phase data X10) and the B-phase input signal Y (B-phase data Y11) that are assumed in the first embodiment are each represented by Expression 1. However, as described above, the A-phase data X10 and the B-phase data Y11 are input as absolute values, so that only positive values are actually taken.

In the first embodiment, up to the third harmonic component is taken into consideration as shown in Formula 1, and a1 in Formula 1 is the coefficient (strength) of the fundamental wave component, and a3 is the coefficient (strength) of the third harmonic component. The same cases are applied to the A phase and the B phase, respectively. Here, the arctangent calculation is performed. Specifically, in this embodiment, the rotation angle θ that satisfies the relational expression shown in Expression 2 obtained from Expression 1 is obtained by the CORDIC rotation calculation method.

Reference numeral 41 is intensity information (a1, a3), which is input as known information from the upper control means to the rotation direction determination unit 401. This intensity information is obtained, for example, by a method in which the encoder signal is frequency-analyzed in advance by FFT (Fast Fourier Transform) and the intensity components of the fundamental wave and the third harmonic are extracted. However, the intensity component does not necessarily have to be the intensity component, and an arbitrary value that reduces the detection error may be selected as the value corresponding to the intensity information.

(Θ rotation calculation)
In the θ rotation calculation, the rotation calculation is performed at the basic rotation angle θ, which is a rotation calculation process for obtaining the coordinate data corresponding to the fundamental wave component. The selection unit 402 that constitutes the θ rotation calculation selects a fixed value (1, 0) of the reference coordinate data 42 at the first rotation calculation in the series of CORDIC calculations, and the CORDIC rotation calculation result after the second calculation that is repeated calculation. Θ rotation coordinate data 48 is selected. When the A-phase data X10 and the B-phase data Y11 are updated and a new arctangent calculation is performed, the above repeated calculation is cleared. Then, a new CORDIC calculation is started, the reference coordinate data 42(1,0) is recursively selected as the first rotation calculation, and thereafter the rotation coordinate data 48 is selected.

The θi rotation unit 404 is a rotation calculation circuit that rotates the θ rotation selection coordinate data 44 in the rotation direction δi 47 by the rotation angle of θi. This rotation calculation has a characteristic of CORDIC rotation calculation, and the rotation calculation shown in Formula 3 is executed with respect to the number i of repeated rotation calculations.

Here, (Xi−1, Yi−1) is the input coordinate data of the θi rotation unit 404, and (Xi, Yi) is the output coordinate data of the θi rotation unit 404. Further, the number of repeated rotation calculations i takes a value from 0 to m−1 with respect to the final number of repeated calculations m. When the number i of repeated rotation calculations is 0 in Equation 3, the coordinate data on the right side is (X-1, Y-1), which is the reference coordinates selected by the selection unit 402 as the first rotation calculation. It is the data 42 (1, 0). Further, the rotation direction δi47 is fixed to 1 (forward rotation (counterclockwise)) when the number of times of repeated rotation calculation i is 0, and thereafter 1 or -1 (reverse rotation (clockwise rotation) depending on the determination result of the rotation direction determination unit 401. )) is selected. Furthermore, the actual rotation angle θi of the rotation calculation with respect to the number of rotation calculations i shown in Formula 3 is shown in Formula 4.

The 2θ rotation correction unit 406 is a correction circuit that corrects an absolute value shift due to a difference in the number of rotation calculations between the θi rotation unit 404 and a 3θi rotation unit 405 described later. Specifically, the calculation shown in Formula 5 is performed.

However, Xin and Yin on the right side are the uncorrected rotational coordinate data 46 output from the θi rotation unit 404, and Xout and Yout on the left side are the rotational coordinate data 48 corrected (updated) by the 2θ rotation correction unit 406. (COSθ, SINθ).

However, this calculation does not necessarily have to be executed by this rotation correction unit, and a calculation circuit that realizes the same function may be arranged in the rotation direction determination unit 401 described below.

The above series of operations corresponds to one repetitive rotation calculation in the basic CORDIC calculation (θ rotation calculation). In the present embodiment, simultaneously with this basic CORDIC calculation (θ rotation calculation), the 3θ rotation calculation shown in FIG. 4 is also executed.

(3θ rotation calculation)
In the 3θ rotation calculation, a rotation calculation is performed at a rotation angle of 3θ, which is a rotation calculation process for obtaining coordinate data corresponding to the third harmonic component. The selection unit 403 that constitutes the 3θ rotation calculation selects the reference coordinate data 43 (1, 0) for the first rotation calculation in the series of CORDIC calculations (3θ), and the CORDIC rotation calculation results for the second and subsequent calculations that are repeated calculations. 3θ rotation coordinate data 49 is selected. When the A-phase data X10 and the B-phase data Y11 are updated and a new arctangent calculation is performed, the above repeated calculation is cleared. Then, a new CORDIC calculation is started, the reference coordinate data 43 (1, 0) is selected again as the first rotation calculation, and thereafter, the rotation coordinate data 49 is selected.

The 3θi rotation unit 405 rotates the 3θ rotation selection coordinate data 45 in the rotation direction δi 47 by the rotation angle of 3θi. This rotation calculation has a characteristic of CORDIC rotation calculation, and the rotation calculation shown in Formula 6 is executed for the repeated rotation calculation number i.

Here, (Xi-1′, Yi-1′) is input coordinate data of the 3θi rotation unit 405, and (Xi′, Yi′) is output coordinate data of the 3θi rotation unit 405 (rotation coordinate data 49 (COS3θ, SIN3θ). )). When the number i of repeated rotation calculations is 0 in Expression 3, the coordinate data on the right side is (X-1', Y-1'), which is selected by the selection unit 403 as the first rotation calculation. It is the reference coordinate data 43 (1, 0). Further, the actual rotation angle 3θi of the rotation calculation with respect to the rotation calculation number i shown in Expression 6 is shown in Expression 7.

The above series of operations corresponds to one repetitive rotation calculation in the CORDIC calculation (3θ rotation calculation), which is one of the features of the present invention. The rotational coordinate data 49 (COS3θ, SIN3θ) obtained here corresponds to the third-order harmonic component included in the encoder signal.

As described above, the rotational coordinate data 48 (COSθ, SINθ) and the rotational coordinate data 49 (COS3θ, SIN3θ) are obtained by the θ rotation calculation and the 3θ rotation calculation. However, in reality, due to the characteristics of the CORDIC rotation calculation, the absolute value increases every time the rotation calculation is executed, so that a certain coefficient is applied to the rotation coordinate data 48 and the rotation coordinate data 49. However, the 2θ rotation correction unit 406 described above is in a state in which the coefficients of the mutual coordinate data are aligned. Therefore, since the calculation result in the rotation direction determination unit described below is not directly affected, the description of the coefficient is omitted. Although the reference coordinate data 42 and 43 are set to (1, 0) in the above, arbitrary coordinate data may be taken as long as the sine component is zero and all cosine components are common. For example, it may be (2,0) or (0.5,0).

(Determination of rotation direction)
The following data is input to the rotation direction determination unit 401. The rotation coordinate data 48 (COSθ, SINθ) and the rotation coordinate data 49 (COS3θ, SIN3θ) obtained by the θ rotation calculation and the 3θ rotation calculation are shown. Further, strength information 41 (a1 and a3) given from the higher-order control means and A-phase data X10 and B-phase data Y11 which are encoder signal inputs are also inputted. In the direction determining unit, the rotation direction δi47 for the next repeated rotation calculation is determined by the calculation formula (determination condition) for determining the direction. Equation 8 shows the discriminant.

When the formula 8 is true, it is a positive rotation (counterclockwise rotation), and δi is 1, and when the formula 8 is false, it is a reverse rotation (clockwise rotation), and δi has a value of -1. However, since the formula 8 can be mathematically derived, the discriminant is not limited to the formula 8. For example, as shown in Expression 9, a discriminant in which the denominator is paid and the division is omitted may be used. Any formula that can be derived while preserving the mathematical meaning of Formula 8 may be adopted in this embodiment.

In this way, the third harmonic component (COS3θ, SIN3θ) is included in the discriminant, and this has the effect of correcting the third harmonic. By determining the rotation direction δi47 as described above, the next repeated rotation calculation can be executed.

(Interpolation position calculation)
By executing the above θ rotation calculation, 3θ rotation calculation, and rotation direction determination m times, which is the final number of repeated calculations, the rotation coordinate data 48 and 49 that satisfy Expression 2 and the δi data string (data) that becomes rotation angle information. History) is obtained. As the final calculation step, the declination calculation unit 407 performs the calculation shown in Formula 10 on the δi data string to obtain the declination θ formed by the rotational coordinate data 48, that is, the interpolation position 12 (where the harmonic component is removed. θ) is required.

As described above, according to the present embodiment, it is possible to calculate the deviation angle (inverse tangent operation) by excluding the harmonic components included in the encoder output signal up to the third order. As a result, it is possible to obtain a highly accurate detection position in which the detection error due to the third harmonic component is removed (corrected). FIG. 15 shows the calculation result in this embodiment when the encoder signal shown in FIG. 13 is input. The calculation result is for the case where the fundamental wave component intensity a1 is 1, and the third harmonic component intensity a3 is -0.094. As shown in the figure, the detection error is about ±0.15 [deg], which is significantly smaller than the case without correction. As a result, the detected displacement amount substantially matches the ideal displacement amount. (Overlapping on the graph)

As described above, the calculation of the third harmonic component (3θ rotation calculation) is performed at the same time as the conventional CORDIC calculation (corresponding to θ rotation calculation). The direction discrimination is also improved by a discriminant formula in which the third harmonic component is added to the conventional CORDIC calculation, and the calculation processing speed is compared with the conventional CORDIC calculation (in the case of only θ rotation calculation). It is not inferior. It is also possible to suppress an increase in the memory for correcting the detection error. The present embodiment can be applied to other forms. For example, in the above description, the rotary motor is used as the movable portion, but a linear motion mechanism may be used instead of this. Further, the driving means may be driven by an actuator such as a motor or a piezo.

(Second embodiment)
In the second embodiment, another configuration of the arctangent calculation unit 304 will be described. Specifically, the configuration is such that the fifth harmonic component is taken into consideration, and a rotation calculation of 5θ is added. Also, the direction discriminant is different. In the following description, the points different from the first embodiment will be mainly described.

(Structure of arctangent calculation unit considering harmonic components)
FIG. 5 shows a configuration diagram of the arctangent calculation unit 304 using the CORDIC rotation calculation capable of correcting up to the fifth harmonic distortion in the present embodiment. The A-phase data X10 and the B-phase data Y11 that are assumed in the second embodiment are each represented by Expression 11. However, as described above, the A-phase data X10 and the B-phase data Y11 are input as absolute values, so that only positive values are actually taken.

In the present embodiment, up to the fifth harmonic component is taken into consideration as shown in Formula 11, and in the Formula 1, a1 is the intensity of the fundamental wave component, a3 is the intensity of the third harmonic component, and a5 is the fifth harmonic component. The wave components are shown, and the cases where the A phase and the B phase are the same are targeted. Here, the arctangent calculation is performed. Specifically, in this embodiment, θ that satisfies the relational expression shown in Expression 12 obtained from Expression 11 is obtained by the CORDIC rotation calculation method.

Reference numeral 51 is intensity information (a1, a3, a5), which is input as known information from the higher-order control means to the rotation direction determination unit 501. This intensity information can be obtained by, for example, performing a frequency analysis on the encoder signal by FFT (Fast Fourier Transform) in advance and extracting intensity components of the fundamental wave, the third harmonic wave, and the fifth harmonic wave.

(Θ rotation calculation)
The basic operation is the same as that of the first embodiment, but the 4θ rotation correction unit 504 is different.

The 4θ rotation correction unit 504 corrects the absolute value shift due to the difference in the number of rotation calculations between the θi rotation unit 404 and a 5θi rotation unit 503 described later. Specifically, the calculation shown in Expression 13 is performed.

However, Xin and Yin on the right side are the uncorrected rotational coordinate data 46 output from the θi rotation unit 404, and Xout and Yout on the left side are rotational coordinate data 48 (corrected (updated) by the 4θ rotation correction unit 504. COS θ, SIN θ).

(3θ rotation calculation)
The basic operation is the same as that of the first embodiment, but a 2θ rotation correction unit 505 is added.

The 2θ rotation correction unit 505 corrects the absolute value shift caused by the difference in the number of rotation calculations between the 3θi rotation unit 405 and a 5θi rotation unit 503 described later. Specifically, the calculation shown in Formula 14 is performed.

However, Xin and Yin on the right side are the uncorrected rotation coordinate data 52 output from the θi rotation unit 405, and Xout and Yout on the left side are the rotation coordinate data 49 (corrected (updated) by the 2θ rotation correction unit 505. COS3θ, SIN3θ).

(5θ rotation calculation)
The selection unit 502 that constitutes the 5θ rotation calculation selects the reference coordinate data 53 (1, 0) for the first rotation calculation in the series of CORDIC calculations (5θ), and the CORDIC rotation calculation results for the second and subsequent calculations that are repeated calculations. The 5θ rotation coordinate data 55 is selected. When the A-phase data X10 and the B-phase data Y11 are updated and a new arctangent calculation is performed, the above repeated calculation is cleared. Then, a new CORDIC calculation is started, the reference coordinate data 53(1,0) is selected again as the first rotation calculation, and thereafter, the rotation coordinate data 55 is selected.

The 5θi rotation unit 503 rotates the 5θ rotation selection coordinate data 54 in the rotation direction δi47 by a rotation angle of 5θi. This rotation calculation has a feature of CORDIC rotation calculation, and the rotation calculation shown in Expression 15 is executed for the repeated rotation calculation number i.

Here, (Xi−1″, Yi−1″) is input coordinate data of the 5θi rotation unit 503. (Xi″, Yi″) represents output coordinate data (rotational coordinate data 55 (COS5θ, SIN5θ)) of the 5θi rotation unit 503. Further, in the equation 14, when the number i of repeated rotation calculations is 0, the coordinate data on the right side becomes (X−1″, Y−1″), which is selected by the selection unit 502 as the first rotation calculation. Reference coordinate data 53 (1, 0). Furthermore, the actual rotation angle 5θi of the rotation calculation with respect to the rotation calculation number i shown in Expression 15 is shown in Expression 16.

The above series of operations corresponds to one time of the repeated rotation calculation in the CORDIC calculation of the 5θ rotation calculation in the present embodiment. Here, the obtained rotational coordinate data 55 (COS5θ, SIN5θ) corresponds to the fifth harmonic component included in the encoder signal.

As described above, the rotation coordinate data 48 (COSθ, SINθ), the rotation coordinate data 49 (COS3θ, SIN3θ), and the rotation coordinate data 55 (COS5θ, SIN5θ) are obtained by the θ rotation calculation, the 3θ rotation calculation, and the 5θ rotation calculation. Was given. However, in reality, due to the characteristics of the CORDIC rotation calculation, the absolute value becomes larger each time the rotation calculation is executed, so that the rotation coordinate data 48, the rotation coordinate data 49, and the rotation coordinate data 55 have coefficients. However, the 4θ rotation correction unit 504 and the 2θ rotation correction unit 505, which have already been described, are in a state in which the coefficients of the mutual coordinate data are aligned. Therefore, since the calculation result in the rotation direction determination unit described below is not directly affected, the description of the coefficient is omitted.

(Rotation direction discrimination)
The rotation direction determination unit 501 includes rotation coordinate data 48 (COSθ, SINθ), rotation coordinate data 49 (COS3θ, SIN3θ), and rotation coordinate data 55 (COS5θ) obtained by θ rotation calculation, 3θ rotation calculation, and 5θ rotation calculation. , SIN5θ) is input. Further, the intensity information 51 (a1 and a3 and a5) given from the higher-order control means and the A-phase data X10 and the B-phase data Y11 which are encoder signal inputs are also inputted. The direction discriminating unit determines the rotation direction δi47 for the next repetitive rotation calculation based on the equation for discriminating the direction. Equation 17 shows the discriminant.

When Expression 17 is true, it is a positive rotation (counterclockwise rotation), δi is 1, and when Expression 16 is false, it is a reverse rotation (clockwise rotation), and δi has a value of -1. However, since the mathematical expression 17 can be derived mathematically as in the first embodiment, the discriminant is not limited to the mathematical expression 17. Any formula that can be derived while preserving the mathematical meaning may be adopted in the present invention.

In this way, the third-order harmonic component (COS3θ, SIN3θ) and the fifth-order harmonic component (COS5θ, SIN5θ) are included in the discriminant, and this constitutes the correction effect up to the fifth-order harmonic.

By determining the rotation direction δi47 as described above, the next repeated rotation calculation can be executed.

(Interpolation position calculation)
By performing the above θ rotation calculation, 3θ rotation calculation, 5θ rotation calculation, and rotation direction determination a final number of times m of calculation, rotational coordinate data 48, 49, 55 satisfying Expression 11 is obtained. As the final calculation step, the declination calculation unit 407 calculates the interpolation position 12 (θ) by performing the calculation shown in Formula 10.

As described above, according to the second embodiment, it is possible to perform arctangent calculation by excluding the harmonic components included in the encoder output signal up to the fifth order. As a result, the detection error due to the third and fifth harmonic components is removed (corrected), and it is possible to obtain a detection position with higher accuracy than in the first embodiment. Similar to the first embodiment, the calculation of the third harmonic component (3θ rotation calculation) and the calculation of the fifth harmonic component (5θ rotation calculation) are performed simultaneously with the conventional CORDIC calculation (corresponding to the θ rotation calculation). It is done. The direction discrimination is improved by a discriminant equation in which the third harmonic component and the fifth harmonic component are added to the conventional CORDIC calculation, and the calculation processing speed is the same as the conventional CORDIC calculation (only θ rotation calculation is performed. If you compare). FIG. 16 shows the calculation result in the second embodiment when the encoder signal shown in FIG. 13 is input. The calculation result is for the case where the fundamental wave component intensity a1 is 1, the third harmonic component intensity a3 is −0.094, and the fifth harmonic component intensity a5 is −0.0023. The detection error is about ±0.02 [deg], which is smaller than that when correcting up to the third harmonic of the first embodiment.

Further, although the correction method up to the 5th harmonic is shown in the second embodiment, the cosine component COS(nθ) and the sine component ( nθ) (n is a positive and odd number) harmonic component can be corrected. Specifically, it is addition up to nθ rotation calculation, addition of a rotation correction unit that corrects the number of rotation calculation times, and change of the discriminant in the rotation direction discrimination unit. As described above, according to the present invention, it is possible to easily provide an arctangent calculation capable of removing higher-order harmonic components.

(Third Embodiment)
In the third embodiment, another configuration of the arctangent calculation unit 304 will be described. The difference from the first embodiment is that the arctangent calculation is provided according to the case where the intensities of the fundamental wave and the harmonic components are different between the A phase and the B phase. In the first embodiment, as shown in Formula 1, the fundamental wave component intensity and the harmonic component intensity in the A phase and the B phase are the same. However, the characteristics of the encoder, the filter processing of the encoder signal, the phase difference correction processing, and the like may be affected, and there is a possibility that they will not be the same. Further, since the correction accuracy of the harmonic component also depends on the accuracy of this intensity information, it is convenient to set the intensity separately for the A phase and the B phase in order to give the intensity with higher accuracy. Good. Therefore, in the present embodiment, the arctangent calculation is provided according to the case where the intensities of the fundamental wave and the harmonic components are different between the A phase and the B phase. In the following description, the points different from the first embodiment will be mainly described.

(Structure of arctangent calculation unit considering harmonic components)
FIG. 6 shows a configuration diagram of the arctangent calculation unit 304 using the CORDIC rotation calculation capable of correcting up to the third harmonic distortion in the present embodiment. The A-phase data X10 and the B-phase data Y11 that are assumed in the third embodiment are each expressed by Expression 17. However, as described above, the A-phase data X10 and the B-phase data Y11 are input as absolute values, so that only positive values are actually taken.

In the present embodiment, the respective component intensities are different between the A phase and the B phase as shown in Expression 18. Therefore, in the present embodiment, θ that satisfies the relational expression shown in Expression 19 obtained from Expression 18 is obtained by the CORDIC rotation calculation method. (Perform arctangent operation).

Reference numeral 61 is intensity information (a1, a3, b1, b3), which is input as known information from the higher-order control means to the rotation direction determination unit 601. This intensity information is obtained, for example, by a method of performing frequency analysis on the encoder signal by FFT (Fast Fourier Transform) in advance and extracting each intensity component.

(Θ rotation calculation)
The θ rotation calculation and the 3θ rotation calculation are the same as those in the first embodiment.

(Rotation direction discrimination)
The rotation direction determination unit 601 receives the rotation coordinate data 48 (COSθ, SINθ) and the rotation coordinate data 49 (COS3θ, SIN3θ) obtained by the θ rotation calculation and the 3θ rotation calculation. Further, the strength information 61 (a1, a3 and b1, b3) given from the higher-order control means, and the A-phase data X10 and the B-phase data Y11 which are encoder signal inputs are also input to the rotation direction determination unit 601. The direction discriminating unit determines the rotation direction δi47 for the next repetitive rotation calculation based on the equation for discriminating the direction. The discriminant is shown in Formula 20.

When Expression 20 is true, it is a positive rotation (counterclockwise rotation), and δi is 1, and when Expression 20 is false, it is a reverse rotation (clockwise rotation), and δi has a value of -1. However, since the mathematical formula 20 can be derived mathematically as in the first embodiment, the discriminant is not limited to the mathematical formula 20. All expressions that can be derived while preserving the mathematical meaning may be adopted in this embodiment. As described above, the rotation direction δi47 is determined, whereby the rotation calculation is repeatedly performed.

As described above, according to the present embodiment, it is possible to provide the arctangent calculation in which the third harmonic component is removed even when the intensities of the fundamental wave and the harmonic components are different between the A phase and the B phase.

FIG. 17 shows the calculation result without correction when the encoder signal shown in FIG. 18 is input, and FIG. 19 shows the calculation result in the third embodiment. In FIG. 19, the A-phase fundamental wave component intensity a1 is 0.94, the B-phase fundamental wave component intensity b1 is 0.90, the A-phase third harmonic component intensity a3 is 0.055, and the B-phase third harmonic component. The calculation result is obtained when the intensity b3 is 0.11. In the case of no correction, the detection error was about ±8 [deg], whereas in the third embodiment, the detection error is significantly reduced to about ±0.2 [deg]. As a result, the detected displacement amount substantially matches the ideal displacement amount. (Overlapping on the graph)
In the present embodiment, by changing the discriminant of the rotation direction discriminator according to the theoretical formula of the assumed encoder signal, it is possible to easily cope with it. Further, also in the present embodiment, it is possible to extend to a configuration capable of removing high-order harmonic components, as in the second embodiment. The extension method is as described in the second embodiment, and the description of the high-order harmonic component removal configuration for the third embodiment is omitted.

(Fourth Embodiment)
In the fourth embodiment, another configuration of the arctangent calculation unit 304 will be described. The configuration is for reducing the calculation load in the rotation direction determination unit. The rotation direction determination process based on the discriminant is a part of the CORDIC rotation calculation and is repeatedly executed. Therefore, it is possible to reduce the calculation load in the rotation direction determination unit, specifically, only add and subtract, and as a result, it is possible to improve the calculation speed of the arctangent calculation. In the following description, differences from the first and third embodiments will be mainly described.

(Structure of arctangent calculation unit considering harmonic components)
FIG. 7 shows a block diagram of the arctangent calculation unit 304 using the CORDIC rotation calculation capable of correcting up to the third harmonic distortion in this embodiment. The A-phase data X10 and the B-phase data Y11 assumed in the fourth embodiment are the same as those in the third embodiment and are as shown in Expression 17. Therefore, in the present embodiment, as in the third embodiment, θ that satisfies the relational expression shown in Expression 18 is obtained by the CORDIC rotation calculation method. (Perform arctangent calculation)
(Generate initial value for rotation calculation)
The configuration of the initial value generation unit 701 is significantly different from the other embodiments. The initial value generation unit 701 is supplied with intensity information 61 (a1, a3 and b1, b3) given by a higher-order control means and A-phase data X10 and B-phase data Y11 which are encoder signal inputs. The initial value generation unit 701 generates four coordinate data based on these pieces of information and supplies them to the rotation calculation in the subsequent stage. Specifically, the initial coordinate data 71 (Ya1, 0), the initial coordinate data 72 (Xb1, 0), the initial coordinate data 73 (Ya3, 0), and the initial coordinate data 74 (Xb3, 0).

(Θ rotation calculation)
Unlike the other embodiments, two θ rotation calculations, a θ rotation calculation 1 and a θ rotation calculation 2, are configured. The selection unit 702 configuring the θ rotation calculation 1 selects the initial coordinate data 71 (Ya1, 0) in the first rotation calculation in the series of CORDIC calculations, and the second and subsequent calculations that are repeated calculations are the CORDIC rotation calculation results. The θ rotation coordinate data 81 is selected. When the A-phase data X10 and the B-phase data Y11 are updated and a new arctangent calculation is performed, the above repeated calculation is cleared. Then, a new CORDIC calculation is started, the updated initial coordinate data 71 (Ya1, 0) is selected again as the first rotation calculation, and thereafter, the rotary coordinate data 81 is selected. The θi rotation unit 706 rotates the θ rotation selection coordinate data 75 in the rotation direction δi47 by the rotation angle of θi. This rotation calculation has a characteristic of CORDIC rotation calculation, and the rotation calculation shown in Expression 3 is executed for the repeated rotation calculation number i. Here, (Xi−1, Yi−1) is the input coordinate data of the θi rotation unit 706, and (Xi, Yi) is the output coordinate data of the θi rotation unit 706. Further, the number of repeated rotation calculations i takes a value from 0 to m−1 with respect to the final number m of repeated calculations. Further, when the number i of repeated rotation calculations is 0 in Expression 3, the coordinate data on the right side is (X-1, Y-1), which is the initial coordinates selected by the selection unit 702 as the first rotation calculation. It is the data 71 (Ya1, 0). Further, the rotation direction δi47 is fixed to 1 (forward rotation (counterclockwise)) when the number of times of repeated rotation calculation i is 0, and thereafter 1 or -1 (reverse rotation (clockwise rotation) depending on the determination result of the rotation direction determination unit 712. )) is selected. The 2θ rotation correction unit 710 corrects the absolute value shift due to the difference in the number of rotation calculations between the θi rotation unit 706 and the 3θi rotation unit 708 and the 3θi rotation unit 709 described later. Specifically, the calculation shown in Expression 5 is performed.

However, Xi and Yi on the right side are the uncorrected rotational coordinate data 79 output from the θi rotation unit 706, and Xi and Yi on the left side are the rotational coordinate data 81 (corrected (updated) by the 2θ rotation correction unit 710. Ya1COSθ, Ya1SINθ). The above series of operations corresponds to one repetitive rotation calculation of the θ rotation calculation 1 in the present embodiment. Here, as a result of the initial coordinate data 71 being given as an initial value, rotational coordinate data 81 (Ya1COSθ, Ya1SINθ) is obtained. The cosine component Ya1COSθ is a calculation result corresponding to one of the four terms of the discriminant used in the rotation direction discriminating unit 712 described later. The same applies to other rotation calculations (θ rotation calculation 2, 3θ rotation calculation 1, 3θ rotation calculation 2). The rotation selection coordinate data 76, the rotation selection coordinate data 77, and the rotation selection coordinate data 78 are selected by the selection unit 703, the selection unit 704, and the selection unit 705, respectively. The rotation unit 709 calculates the rotation. The pre-correction rotational coordinate data 80 that has been rotationally calculated by the θi rotation unit 707 is corrected by the 2θ rotation correction unit 711 to obtain rotational coordinate data 82. As described above, rotational coordinate data 82 (Xb1COSθ, Xb1SINθ), rotational coordinate data 83 (Ya3COSθ, Ya3SINθ), and rotational coordinate data 84 (Xb3COSθ, Xb3SINθ) are obtained, respectively.

(Rotation direction discrimination)
An arithmetic expression for the direction determination by the rotation direction determination unit 712 in the present embodiment is represented by Formula 21. This is a form in which the denominator of Expression 20 is paid and the division is omitted.

As in the other embodiments, when the expression 21 is true, it is a positive rotation (counterclockwise rotation), δi is 1, and when the expression 20 is false, it is a reverse rotation (clockwise rotation), and δi has a value of −1. A feature of the present embodiment is that the four terms on the left side shown in Expression 21 are included in the results (rotational coordinate data 81, 82, 83, 84) of each rotation calculation in the preceding stage. Therefore, the rotation direction discriminating unit 712 can determine the rotation direction δi47 only by performing addition and subtraction and determining whether it is positive or negative. This means that the harmonic distortion correction CORDIC rotation processing unit is configured without a multiplier. Therefore, the speed of the repetitive rotation calculation unit is increased.

As described above, according to the present embodiment, it is possible to configure the CORDIC rotation calculation processing section without a multiplier. In addition, it is possible to perform arctangent calculation by removing harmonic components included in the encoder output signal up to the third order, and as a result, it is possible to obtain a highly accurate detection position in which a detection error due to the third order harmonic component is removed (corrected). The calculation result in this embodiment is the same as that in the third embodiment and is as shown in FIG. Further, also in the present embodiment, it is possible to extend to a configuration capable of removing high-order harmonic components as in the second embodiment. The extension method is as described in the second embodiment, and the description of the high-order harmonic component removal configuration for this embodiment is omitted.

(Fifth Embodiment)
The fifth embodiment shows an application example as a harmonic distortion removal filter in the CORDIC rotation calculation in consideration of the harmonic component described in the other embodiments.

FIG. 8 is a diagram showing an example of the configuration of the displacement detector in the third embodiment.

The arctangent calculation unit 801 has an arctangent calculation function, unlike the configurations described in the other embodiments, but does not have a function of removing harmonic components. This is a configuration of arctangent calculation by a normal CORDIC. A major difference from the first embodiment is that it has a harmonic wave removing filter unit 802. In this embodiment, since the arc tangent calculation unit 801 does not calculate the harmonic component, the preceding harmonic removal filter unit 802 performs the process of removing the harmonic component.

FIG. 9 shows a block diagram of the harmonic elimination filter unit 802. The harmonic removal filter unit 802 has substantially the same configuration as the arctangent calculation unit 304 of the first embodiment. The basic operation is the same as that described in the first embodiment. The change is that the declination calculation unit 407 is removed and the rotation coordinate data 48 (COSθ, SINθ) which is the calculation result of the θ rotation calculation is output instead of the rotation angle output. .. The harmonic component removal X data 14 corresponds to the cosine component (COSθ) of the rotational coordinate data 48, and the harmonic component removal Y data 16 corresponds to the sine component (SINθ) of the rotational coordinate data 48. As a result of a series of repetitive rotation calculations in this embodiment, rotational coordinate data 48 (COSθ, SINθ) that is a fundamental wave component and rotational coordinate data 49 (COS3θ, SIN3θ) that is a third-order harmonic component are additionally obtained. You can Using this effect, it is a feature of this embodiment that only the fundamental wave component is extracted, that is, it functions as a signal filter that removes the third harmonic component.

As described above, according to this embodiment, it is possible to provide the information processing device of the encoder having the filter function of removing the third harmonic component contained in the A-phase and B-phase signals. FIG. 20 shows the result of signal processing of the encoder signal including the harmonic components shown in FIG. 14 in the fifth embodiment. As shown in FIG. 20, it can be seen that the third harmonic component is removed and a highly accurate signal of only the fundamental component is obtained. Also in this embodiment, it is possible to take the detailed configuration of the other embodiments. Specifically, a configuration capable of removing higher harmonic components as in the second embodiment, a configuration in which the intensities of the fundamental wave and the harmonic components are different between the A phase and the B phase as in the third embodiment, and Similar to the fourth embodiment, it has a multiplication-free configuration. These are as described in the description of each embodiment, and detailed description will be omitted.

(Other embodiments)
Since the arctangent calculation is performed based on the CORDIC calculation algorithm as the basic principle in the above embodiment, the embodiment in which the CORDIC calculation algorithm is also used for the calculation of the harmonic component has been described. However, even if another method is used for the calculation of the harmonic component, it does not depart from the gist of the present invention. For example, it is possible to convert the basic rotation calculation result (rotation coordinate data by the θ rotation calculation) obtained by the CORDIC rotation calculation into the rotation coordinate data obtained by the 3θ rotation calculation using the trigonometric triple angle formula. You may input these rotation coordinate data into a direction determination part, and determine the next rotation calculation direction.

In addition, in the above-described embodiment, a recursive processing type configuration is adopted in which a single rotation unit performs repeated rotation calculation processing operations a plurality of times. However, it is also possible to realize a pipeline processing type configuration in which a plurality of stages of rotation units are provided in the same rotation calculation unit, and these are connected in a pipeline to perform a rotation calculation processing operation. The above-mentioned motor control device can be used for a pan head such as a stage of a network camera or a manufacturing apparatus to improve the smoothness of operation of the pan head. The network camera includes a CPU, a ROM, a RAM, an image pickup unit, a platform, and a network interface.

Further, in the above embodiment, the case where the two-phase encoder signal for detecting the position includes the harmonic component has been described. However, the present invention is not limited to encoder signals, and can be applied to all two-phase signals including harmonic components.

For example, a two-phase signal corresponding to the motor rotation angle can be detected from the drive current when driving a three-phase brushless motor. That is, the motor rotation angle can be calculated from the detected two-phase signal by ATAN processing.

It is known that a harmonic component is superimposed on this two-phase signal. Therefore, by applying this embodiment, it is possible to remove the influence of the harmonic components of the two-phase signal and detect the accurate motor rotation angle.

The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or device via a network or various storage media, and the computer (or CPU, MPU, etc.) of the system or device reads the program. This is the process to be executed.

401 rotation direction determination unit 402 selection unit (θ rotation calculation)
403 Selector (3θ rotation calculation)
404 θi rotation unit 405 3 θi rotation unit 406 2θ rotation correction unit 407 declination calculation unit

Claims (11)

Acquisition means for acquiring the analog wavy signal of the first phase and the analog wavy signal of the second phase obtained by measuring the moving body,
Using a first cosine wave and a second cosine wave that is a higher harmonic than the first cosine wave for the first phase analog wave signal, and for the second phase analog wave signal Phase information satisfying the arctangent relational expression using the first sinusoidal wave and the second sinusoidal wave that is a higher harmonic than the first sinusoidal wave is included in the arctangent relational expression by iterative calculation. An information processing device comprising: a calculating unit that calculates by converging a rotation angle. The information processing apparatus according to claim 1, wherein the iterative calculation is a CORDIC operation. 3. The information processing apparatus according to claim 1, wherein in the iterative calculation, initial values of the first sine wave and the second sine wave are zero. The information processing apparatus according to claim 2, wherein the calculation unit has a correction unit that corrects an absolute value of coordinate data in the CORDIC calculation. The calculating means has a plurality of arithmetic circuits for performing respective operations included in the iterative calculation,
The information processing apparatus according to any one of claims 1 to 4, wherein the plurality of arithmetic circuits are connected in a pipeline. The calculating means has a plurality of arithmetic circuits for performing respective operations included in the iterative calculation,
The information processing device according to claim 1, wherein the plurality of arithmetic circuits are recursively connected. 7. The output means for outputting a signal from which at least one component of the second cosine wave and the second sine wave is removed is further included. The information processing device described. The information processing apparatus according to claim 7, wherein the output unit outputs the signal as an encoder signal indicating a positional displacement of the moving body. 9. The information processing apparatus according to claim 1, further comprising a control unit that controls rotation of a motor that controls the position of the moving body based on the calculation result of the calculation unit. .. Acquisition means, an acquisition step of acquiring the analog wavy signal of the first phase and the analog wavy signal of the second phase obtained by measuring the moving body,
The calculating step uses a first cosine wave and a second cosine wave that is a higher harmonic than the first cosine wave for the first phase analog wavy signal, and uses the second phase analog wavy signal. The information of the phase satisfying the relational expression of the arctangent using the first sinusoidal wave and the second sinusoidal wave that is a higher harmonic wave than the first sinusoidal wave with respect to the signal is obtained by iterative calculation and the arctangent relationship is obtained. And a calculation step of calculating by converging the rotation angle included in the formula. Computer,
Acquisition means for acquiring the analog wavy signal of the first phase and the analog wavy signal of the second phase obtained by measuring the moving body,
Using a first cosine wave and a second cosine wave that is a higher harmonic than the first cosine wave for the first phase analog wave signal, and for the second phase analog wave signal Phase information satisfying the arctangent relational expression using the first sinusoidal wave and the second sinusoidal wave that is a higher harmonic than the first sinusoidal wave is included in the arctangent relational expression by iterative calculation. A program for causing an information processing apparatus to function, comprising: a calculating unit that calculates by converging a rotation angle.